DESCRIPTION: The proposed research examines differentials in involuntary reproductive loss associated with social structure. Analysis is confined to losses that occur after the second trimester of pregnancy, but that extend through the first year after birth. The investigators propose to use data from the Czech Republic that include all registered terminations of pregnancies and infant deaths registered between 1986 and 1991. The data include live births and infant deaths which will be linked back to birth records, together with spontaneous and induced abortions. The investigators propose to model: 1) risk of death following a live birth, 2) risk of involuntary loss following a conception that survives to the start of the second trimester, and 3) overall risk of loss (either by infant death or fetal loss) for conceptions that survive to the start of the second trimester. The proposal offers an opportunity to document mortality in the very earliest stages of life and to explore variation in that mortality according to differentials in factors that have previously been found to be associated with that variation.